Despite the rapid and massive acclaim Muller received following his presentation revealing X-ray induced gene mutation at the Fifth International Genetics Congress in September, 1927 , his striking gene mutation interpretation was not immune to criticism. In fact, Muller was acutely sensitive to insightful, yet private criticism, by his closest friend Edgar Altenburg  that he had “only” produced hundreds of transgenerational phenotypic changes with no proof of his gene mutation interpretation. While the production of massive numbers of such genetic changes in such a short period of time was overwhelming to many, for Altenburg it remained a potentially important observation in search of an explanation. He told Muller that he needed to show that the high dose mutation-inducing X-rays were not merely blasting holes in large sections of chromosomes, producing the heritable phenotypic changes via a chromosomal deletion rather than a gene mutation mediated process. For Altenburg this was the crucial question and Muller knew that he was correct.
For Muller this criticism was devastating as he had come so close to the goal of gene mutation discovery, only to be “denied” the satisfaction by his best friend, who offered a more objective perspective. While Muller needed to hear this criticism, he also was blinded by ambition to be first on perhaps the most critical question of the day in biology, that is, discovery of the mechanism of evolution.
In the years leading up to the key research of Muller on X-ray induced mutations he had made it clear that finding a means to induce mutation at the level of gene would be a profound advance as it may provide a mechanism that could drive evolution [8, 14]. Muller was dismissive of multiple claims of induced mutation during the early 1920s that were mediated at the level of chromosome rather than gene. Muller was firm that unless the induced changes occurred at the level of gene their significance would be very limited, not a qualitative advance. Altenburg was forcing Muller to be as self-critical as he was critical of numerous previously failed attempts by others to address the same issue.
Muller’s  ground breaking report on July 22, 1927 in the journal Science that he had induced hundreds of gene mutations by X-rays was based upon the production of allomorphic (i.e. visual) transgenerational phenotypic changes and lethality. He made this gene mutation claim/interpretation in an article that discussed his findings, but failed to include any data. The scientific community would have to wait until the second week of September that year at the International Genetics Congress for Muller to present the actual findings .
A review of his Nobel Prize data notebook (Muller, Lilly Library, University of Indiana-Bloomington) reveals that findings were derived from three experiments with the second experiment being wrapped up with respect to data collection in the laboratory at the end of May, 1927. The dates of the third experiment were not recorded in the lab notebook although Muller  indicated that he ran out of time to include a control group, probably in order to prepare for and travel to the Genetics Congress. The data discussed in Science matches most closely to that reported in experiment # 1 that was completed in the laboratory in December, 1926.
While it is unknown when Altenburg precisely offered his criticism to Muller, the actions of Muller to introduce a profound advance on one of the most significant questions confronting biology in the journal Science without any proof was very brash under any circumstances. Permitting this to occur was also similarly flamboyant for the editor, James McKeen Cattell. A careful evaluation of the entire preserved correspondence record of Cattell, as an individual and Science editor did not reveal the arrangement between him and Muller. Likewise, this is the case following a careful review of the Muller correspondence files.
That Muller had the personality to manipulate the scientific community and the general public can be seen in this Nobel Prize Lecture (December 12, 1946) where he deliberately deceived the audience arguing that there was no possibility of a threshold dose response for radiation induced mutations when he had just observed data supporting such a conclusion from Caspari and Stern from the University of Rochester, one month prior to the Nobel Prize Lecture [4, 5]. Muller offered this deception to promote his long held ideology in support of a linear non-threshold dose response for mutation and cancer risks. Muller [24, 25] would also deliberately offer inaccurate and deceptive comments in the scientific literature to promote linear non threshold (LNT), expressly contradicting private correspondence between himself and Stern . Such high stakes risk taking and bluffing behavior of Muller surfaced two decades prior to his Nobel Prize Lecture, when he published a description of these findings without data, knowing that he lacked proof for his key gene mutation conclusion. Muller had the chance to submit this most significant research for peer review but choose not to do so. As pointed out, the conference proceedings paper with the data lacked a methods section, provided no references cited and even failed to acknowledge the prior recent paper (January, 1927) by Gager and Blakeslee  which provided what appears to be the first evidence of radiation induced gene mutation. Until this point in his career, Muller published his papers within journals with a well-established peer review record as he was guided into this scientific acculturation by his Columbia University mentor, Thomas Hunt Morgan. Yet, for the paper of greatest significance he bizarrely published a discussion of preliminary findings with no data and then avoided peer review with a poorly developed paper in the Congress proceedings .
It is likely that Muller was both devastated by the criticism of Altenburg and could not find any way to address it, at least not in the short term. Yet, he knew he was in a race to be first, even if Gager and Bakeslee may have some claim to the honor he could handle them with a series of Muller-like blistering challenges. It appears that Muller rolled the proverbial dice, ignored, at least for the time being, the criticisms of Altenburg, and went on to claim his “unequivocal” belief in having produced “gene” mutations, all the while knowing that he had not proven it. This was similar to his deceptive performance at the Nobel Prize Lecture, showing the same risk taking behavioral trait. Muller had confidence that his duplicity would not be revealed by those that knew of his scamming behaviors, such as Stern and Caspari and perhaps a few others at the University of Rochester for the Nobel Prize Lecture, and Altenburg for his Nobel Prize research.
Muller would find a means to address the criticism of Altenburg but it would take several years. He adopted the concept of reverse mutation and thought that it could provide indirect, but functional proof that his treatments would yield genes that would still be working and therefore not destroyed by the high doses of radiation via massive deletions. However, it is curious to note that Muller mentioned that he had not observed any reverse mutations in the Science paper. Thus, in his own experimental model and methods of mutation induction he had not observed the reverse mutation phenomenon. However, in the fall of 1927 he claimed to have observed two examples [18, 23] and then in early 1928 his colleague at Texas James Patterson found evidence of a third example . These very limited observations offered Muller a glimmer of hope to perhaps address the criticism of Altenburg, but only time and experimental evidence would tell. Thus, even a year after the publication in Science Muller had no proof that he had induced gene mutation and was now going to try to support this interpretation via a collaboration with Patterson. At the December, 1927  American Association for the Advancement of Science (AAAS) Conference in Nashville Muller provided the a glimpse into the emerging importance of the reverse mutation concept as he stated that the newly observed two reverse mutations should provide more information on gene structure than “gene” loses and be more significant in the study of evolution, the central theme of biology. In fact, the argument to support the claim of gene mutation would morph into that of reverse mutation .
This discussion provides context and a plausible explanation for why Muller deliberately avoided peer review. A case can be made that he accepted the validity of Altenburg’s concerns. This would have suggested to him that there was the possibility/probability of this paper being rejected. Muller was caught in an ethical dilemma: he could address the criticisms of Altenburg or finesse his manuscripts and presentations to avoid possible devastating critical reviews while accepting the uncritical acclaim of the scientific and world communities who were overwhelmed and transformed with his capacity to produce transgenerational phenotypic changes; this same community then accepted his gene mutation interpretation based largely on an appeal to authority.
In a September 1928 publication (based on an April 15, 1928 presentation at the National Academy of Sciences) in the Proceedings of the National Academy of Sciences (PNAS) Muller acknowledged the criticisms of Altenburg . He also indicated that Patterson had found an example of a reverse mutation. Muller then stated that he was now going to follow up on the suggestion of his wife, a mathematician, to address the Altenburg criticism via the study of reverse mutation. At this point Muller was seemingly aware of only three cases of what might be X-ray induced reverse mutation in the PNAS paper. It is also curious that Muller never acknowledged the two examples of reverse mutation that he referenced in the AAAS presentation nine months earlier.
Nearly three years after the Altenburg criticism Patterson and Muller  would provide support for the gene mutation interpretation using the reverse gene mutation. On page 574 they concluded that “the above (reverse mutations) demonstrates that mutations can be produced in both of two opposite directions at the same locus.” They further claimed that these findings were “irreconcilable with the view that all mutational changes by X-rays consist of losses.”
In retrospect the original “proof” of gene mutation using alleomorphic phenotypic changes-as both an observation and explanation-was not convincing to Muller following the Altenburg criticism. Thus, overtime Muller came to use the reverse mutation observations as the means to explain the induced transgenerational phenotypic changes. In some ways, this was a type of scientific bait and switch process that was nearly unrecognizable, yet critically important over the issue of gene mutation discovery. Nonetheless, the scientific community was mesmerized by the findings and publicity of Muller to such an extent that it quickly and uncritically came to believe that Muller had produced gene mutations based on an interpretation lacking scientific justification.
While Muller may have thought that the reverse mutation findings may have salvaged the gene mutation interpretation, even though three years too late, it too would be challenged very seriously by another thoughtful equal peer of Muller, the corn geneticist Lewis Stadler at the University of Missouri. In fact, Stadler  had been only a few months behind Muller in the reporting of X-ray induced mutations. While Stadler at first accepted the gene mutation interpretation of Muller this would change some four years later.
In 1929 Barbara McClintock  developed a vastly improved cytogenetic method for assessing the characteristics and appearance of chromosomes of corn. At that time this technique was far more advanced than had been developed for Drosophila . In 1931 , McClintock provided a cytogenetic evaluation of X-ray induced mutations in corn under the direction of Stadler. He had observed transgenerational phenotypic changes similar to those produced by Muller with Drosophila. However, McClintock’s advanced cytogenetic methods revealed that the so called “gene” mutations that Stadler believed he had produced were most likely gross chromosomal deletions. Thus, McClintock concluded that the doses of X-rays used by Stadler were simply punching holes in the chromosomes and that he had not induced gene mutations as he had thought. After some deliberation, Stadler [32, 33] came to accept the new insights of McClintock and soon came to question whether this was also the case with Muller’s gene mutations as first suggested by Altenburg. While Muller claimed that the X-ray induced reverse mutation explanation addressed the question, this position was not convincing to Stadler  and others (see  for a review) who offered a series of experimentally based alternative explanations (e.g. position effect, transpositional elemental effects) for Muller’s reverse mutation phenotype findings without the need for gene mutation).
Overlooked in this debate was the dissertation of Clarence P. Oliver, Muller’s student at the University of Texas. In his dissertation (i.e., Conclusion #10)  and in his journal publication (i.e., Conclusion #11) , Oliver stated “It is not possible to produce gene mutations without producing gene rearrangement, or vice versa”. This statement undercuts Muller’s conclusion that he produced “point mutations” at doses far higher than those used by Oliver . Given this contradictory statement, it is curious that Muller approved this dissertation since this statement was very Stadler-like.
While most in the scientific community seemed to believe that Muller had induced gene mutation, Muller and Stadler would contest this point until Stadler died in 1954 with neither side conclusively winning this debate although the preponderance of the evidence favored Stadler . In the end, Stadler adopted the perspective of Altenburg that Muller mistook an observation (i.e. transgenerational phenotypic changes) for an explanation (i.e., gene mutation). However, once nucleotide analysis methods were developed, he was shown to have achieved his transgenerational phenotypic changes via the “punching of holes” in chromosomes via modest to extraordinarily large deletions rather than via point mutations [8,9,10,11].